Final_Code for Ex#4.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Exercise 4. Ensemble Models\n",
    "#### Use the iris data set from sklearn."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Loading libraries\n",
    "from sklearn.datasets import load_iris\n",
    "from sklearn import model_selection\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.model_selection import cross_val_score\n",
    "from sklearn.decomposition import PCA\n",
    "from sklearn.model_selection import GridSearchCV \n",
    "from sklearn import metrics\n",
    "from sklearn.naive_bayes import GaussianNB\n",
    "from sklearn.metrics import accuracy_score, f1_score, precision_score, recall_score, classification_report, confusion_matrix\n",
    "from sklearn.ensemble import RandomForestClassifier\n",
    "from sklearn.ensemble import ExtraTreesClassifier\n",
    "from sklearn.ensemble import GradientBoostingClassifier\n",
    "from sklearn.datasets import make_moons, make_circles, make_classification\n",
    "from sklearn.gaussian_process.kernels import RBF\n",
    "from sklearn.model_selection import learning_curve\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "from matplotlib.colors import ListedColormap\n",
    "%matplotlib inline\n",
    "import seaborn as sns\n",
    "import warnings\n",
    "warnings.filterwarnings('ignore')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Question 21. Load the iris dataset. What are the features?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>data</th>\n",
       "      <th>target</th>\n",
       "      <th>target_names</th>\n",
       "      <th>DESCR</th>\n",
       "      <th>feature_names</th>\n",
       "      <th>filename</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>[[5.1, 3.5, 1.4, 0.2], [4.9, 3.0, 1.4, 0.2], [...</td>\n",
       "      <td>[0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ...</td>\n",
       "      <td>[setosa, versicolor, virginica]</td>\n",
       "      <td>.. _iris_dataset:\\n\\nIris plants dataset\\n----...</td>\n",
       "      <td>[sepal length (cm), sepal width (cm), petal le...</td>\n",
       "      <td>C:\\Users\\dansa\\Anaconda3\\lib\\site-packages\\skl...</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                data  \\\n",
       "0  [[5.1, 3.5, 1.4, 0.2], [4.9, 3.0, 1.4, 0.2], [...   \n",
       "\n",
       "                                              target  \\\n",
       "0  [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ...   \n",
       "\n",
       "                      target_names  \\\n",
       "0  [setosa, versicolor, virginica]   \n",
       "\n",
       "                                               DESCR  \\\n",
       "0  .. _iris_dataset:\\n\\nIris plants dataset\\n----...   \n",
       "\n",
       "                                       feature_names  \\\n",
       "0  [sepal length (cm), sepal width (cm), petal le...   \n",
       "\n",
       "                                            filename  \n",
       "0  C:\\Users\\dansa\\Anaconda3\\lib\\site-packages\\skl...  "
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Loading dataset\n",
    "\n",
    "data = pd.DataFrame.from_dict(load_iris(), orient = 'index').T\n",
    "data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['data', 'target', 'target_names', 'DESCR', 'feature_names', 'filename'], dtype='object')"
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#column in dataset\n",
    "data.columns"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Question 22. Create a DataFrame of given iris dataset"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "   sepal_length  sepal_width  petal_length  petal_width\n",
      "0           5.1          3.5           1.4          0.2\n",
      "1           4.9          3.0           1.4          0.2\n",
      "2           4.7          3.2           1.3          0.2\n",
      "3           4.6          3.1           1.5          0.2\n",
      "4           5.0          3.6           1.4          0.2\n",
      "        Feature_name\n",
      "0  sepal length (cm)\n",
      "1   sepal width (cm)\n",
      "2  petal length (cm)\n",
      "3   petal width (cm)\n"
     ]
    }
   ],
   "source": [
    "#Loading lists from each column of data1 to indiviual dataframe.\n",
    "df_feature=pd.DataFrame(data.data[0])\n",
    "df_target=pd.DataFrame(data.target[0])\n",
    "df_target_names=pd.DataFrame(data.target_names[0])\n",
    "df_feature_names=pd.DataFrame(data.feature_names[0],columns=['Feature_name'])\n",
    "df_feature.columns=['sepal_length','sepal_width','petal_length','petal_width']\n",
    "print(df_feature.head())\n",
    "print(df_feature_names)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(150, 4)\n",
      "(150, 1)\n"
     ]
    }
   ],
   "source": [
    "#columns and rows for data feature\n",
    "print(df_feature.shape)\n",
    "print(df_target.shape)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Question 23. Split the dataset into training and test sets (30%)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [],
   "source": [
    "data2= load_iris()\n",
    "X=pd.DataFrame(data2.data)\n",
    "Y=pd.DataFrame(data2.target)\n",
    "\n",
    "#Splitting data\n",
    "X_train, X_test, Y_train, Y_test = train_test_split(X, Y, test_size=0.3)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Question 24. Create a Gaussian RandomForestClassifier as clf (2,000 estimators and a depth of 2)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [],
   "source": [
    "rf=RandomForestClassifier(n_estimators=2000, max_depth=2) # for parametes (n_estimators=2000, max_depth=2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [],
   "source": [
    "rf.fit(X_train,Y_train)\n",
    "predicted=rf.predict(X_test)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Question 25. Determine the feature importance. Which one is the most important?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Feature_name\n",
       "petal width (cm)     0.464755\n",
       "petal length (cm)    0.409528\n",
       "sepal length (cm)    0.110643\n",
       "sepal width (cm)     0.015074\n",
       "dtype: float64"
      ]
     },
     "execution_count": 39,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "rf_feature_imp = pd.Series(rf.feature_importances_,index=df_feature_names.Feature_name).sort_values(ascending=False)\n",
    "rf_feature_imp"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "No handles with labels found to put in legend.\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "#Feature importance for Random Forest Model\n",
    "# Creating a bar plot\n",
    "sns.barplot(x=rf_feature_imp, y=rf_feature_imp.index)\n",
    "# Add labels to your graph\n",
    "plt.xlabel('Feature Importance Score')\n",
    "plt.ylabel('Features')\n",
    "plt.title(\"Visualizing Important Features\")\n",
    "plt.legend()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Amongst all four features , it seems \"petal width (cm)\" is most important for Random Forest regression model."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Question 26. Use scikitlearn to determine the accuracy level. What is your assessment?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy of Random forest classfier model: 0.96\n"
     ]
    }
   ],
   "source": [
    "#Check accuracy\n",
    "print(\"Accuracy of Random forest classfier model: %.2f\" % accuracy_score(Y_test,predicted))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Question 27. Use the Gradient Boosting and the ExtraTrees classifiers to fit the models and predict test data"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Gradient Boosting Classifier"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "GradientBoostingClassifier(criterion='friedman_mse', init=None,\n",
       "              learning_rate=0.1, loss='deviance', max_depth=2,\n",
       "              max_features=None, max_leaf_nodes=None,\n",
       "              min_impurity_decrease=0.0, min_impurity_split=None,\n",
       "              min_samples_leaf=1, min_samples_split=2,\n",
       "              min_weight_fraction_leaf=0.0, n_estimators=2000,\n",
       "              n_iter_no_change=None, presort='auto', random_state=None,\n",
       "              subsample=1.0, tol=0.0001, validation_fraction=0.1,\n",
       "              verbose=0, warm_start=False)"
      ]
     },
     "execution_count": 42,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Fitting model\n",
    "gb= GradientBoostingClassifier(n_estimators=2000, max_depth=2)# for parametes (n_estimators=2000, max_depth=2)\n",
    "gb.fit(X_train,Y_train)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [],
   "source": [
    "#making predictions\n",
    "gb_pred=gb.predict(X_test)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Extra trees Classifier"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "ExtraTreesClassifier(bootstrap=False, class_weight=None, criterion='gini',\n",
       "           max_depth=2, max_features='auto', max_leaf_nodes=None,\n",
       "           min_impurity_decrease=0.0, min_impurity_split=None,\n",
       "           min_samples_leaf=1, min_samples_split=2,\n",
       "           min_weight_fraction_leaf=0.0, n_estimators=2000, n_jobs=None,\n",
       "           oob_score=False, random_state=None, verbose=0, warm_start=False)"
      ]
     },
     "execution_count": 44,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Fitting model\n",
    "et= ExtraTreesClassifier(n_estimators=2000, max_depth=2)# for parametes (n_estimators=2000, max_depth=2)\n",
    "et.fit(X_train,Y_train)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [],
   "source": [
    "#making predictions\n",
    "et_pred=et.predict(X_test)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Question 28. Compute the accuracy for the GB and XTrees models"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy of Gradient Boosting classfier model: 0.98\n",
      "Accuracy of Extra trees classfier model: 0.96\n"
     ]
    }
   ],
   "source": [
    "#Computing accuracy\n",
    "print(\"Accuracy of Gradient Boosting classfier model: %.2f\" % accuracy_score(Y_test,gb_pred))\n",
    "print(\"Accuracy of Extra trees classfier model: %.2f\" % accuracy_score(Y_test,et_pred))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Question 29. Provide the feature importance for the GB and XTrees models."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Gradient Boosting\n",
    "gb_feature_imp = pd.Series(gb.feature_importances_,index=df_feature_names.Feature_name).sort_values(ascending=False)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "No handles with labels found to put in legend.\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAc0AAAEWCAYAAAAEvMzxAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDMuMC4yLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvOIA7rQAAIABJREFUeJzt3XmcHVWZ//HPlySQQMKSEFkNzc5ASIIJOOygOCqjgD8jqBkw6sgICm6ADCIybIrwU2dEwYTBgOCIojAQZJeETSQJZkMIa4AIQliyAGEJeeaPOlcqne6+53bf27fT+b5fr/tK3apT5zynbqefPqfqVikiMDMzs+rWanYAZmZmqwsnTTMzs0xOmmZmZpmcNM3MzDI5aZqZmWVy0jQzM8vkpGnWAUkXSfp2g9uYIulf0/I4STdn7HODpM80Mi4zW5WTpq2xJN0k6Yw21h8q6W+S+kbEFyPizO6KKSKuiIh/yij34Yi4tN7tSzpA0oJ619sZklokhaS+daqvat8kTZL0pqRXSq8j6tB2SNquq/VY8zlp2ppsEnCkJLVafyRwRUQs7/6QDKBeibKTvh8RA0uvK5sYCwCS+jQ7Bis4adqa7BpgMLBvZYWkjYCPAJel95MknZWWN5Y0WdIiSS9JulPSWmnbSiOJVvttlPZbKOnltLxlWwFJGi/prrR8UqsRz1uSJqVt5Snd8ZLuknR+qv8JSR8u1bm1pDskLZV0q6SfSLo85wClds6SdE+K4TpJQyRdIWmJpGmSWkrlQ9Lxkh6X9IKk80rHaC1Jp0p6UtLzki6TtEHaVhlVfl7SU8AfgDtStYtS23tK2lbSHyS9mOq/QtKGpfbnSzpB0mxJiyVdKam/pPWAG4DNS8dz85xjUKp7c0m/TZ/jE5KOL23bQ9If08/Gs5IukLR22lbpx6zKyLX8Obc6dtul5UmSLpT0e0mvAgdKWid9xk9Jek7FqYMBqXy7P5tWXz6otsaKiGXAr4GjSqsPBx6KiFlt7PINYAEwFNgEOAXIuQ/lWsDPga2AYcAy4IKM+P4+4gH+AViY4m3Le4F5wMbA94H/Lo2gfwncBwwBTqcYSdfik2mfLYBtgT+m/gwGHgS+06r8x4AxwHuAQ4HPpfXj0+tAYBtgIKseh/0p+vpBYL+0bsN0HP4ICPgusHkq9+7Up7LDgQ8BWwMjgPER8SrwYeCZ0gjymdwDkBLQdcCsdBzeD3xV0gdTkbeBr1Ec/z3T9mMBIqLSj5E1jlw/DZwNDALuAs4FdgBGAdulOE5LZTv7s2k1ctK0Nd2lwCcqf7FTJND2zhW+BWwGbBURb0XEnZFx8+aIeDEifhsRr0XEUopfhPvnBphiuwb4z4j4fTvFnoyIiRHxdop/M2ATScOA3YHTIuLNiLgLuDa37eTnEfFYRCymGK09FhG3punr3wC7tSp/bkS8FBFPAT8CPpXWjwN+EBGPR8QrwL8Dn9TKU7GnR8Sr6Q+aVUTEoxFxS0S8ERELgR+w6rH8r4h4JiJeokh0o2rs7wlpxLZI0gtp3e7A0Ig4Ix3Hx4GJFH9QEBEzIuLeiFgeEfOBn7URV63+NyLujogVwBvAF4CvpWO7FDin0j6d/Nm02jlp2hotJZGFwKGStqH45fjLdoqfBzwK3JymH0/OaUPSupJ+lqYll1BMO26o/PNU/w3Mi4hzOyjzt8pCRLyWFgdSjMheKq0DeDqz3YrnSsvL2ng/sFX5cv1PphhI/z7ZaltfipFRVmyS3iXpV5L+mo7l5RSju7K/lZZfayO+as6PiA3Tq1L3VhRTu5VkuohiNLdJimuHND36txTXOW3EVavysRgKrAvMKLV/Y1oPnfzZtNo5aZoV5y+PopiCvDkinmurUEQsjYhvRMQ2wEeBr0t6f9r8GsUvtYpNS8vfAHYE3hsR6/POtGPrC5BWkX757Qh8vob+lD0LDJZUju3dnawrV7n+YUBlGvQZiuRT3raclZNwtLNc8d20fkQ6lv9CxnHsoL5cTwNPlJLphhExKCIOTtsvBB4Ctk9xnVIlrlcp/bxI2rSNMuV4X6D4A2WXUvsbpKn7aj+bVkdOmmZF0jyIYvqr3a9xSPqIpO3SucIlFOex3k6bZwKfltRH0odYeWpuEMUvvEWSBrPqOcD22vswcDxwWHvTldVExJPAdOB0SWtL2pPil2ojnaji4qd3A18BKufw/gf4mooLkwZSjMau7OAq5YXACorznxWDgFcojuUWwIk1xPUcMKRy8VGN7gOWSPqmpAHpcx4uafdSXEuAVyTtBBzTRtvlfswCdpE0SlJ/Vj0vu5I0RTsR+KGkdwFI2qJyTrXKz6bVkZOmrfHSOah7gPXo+Hzf9sCtFL+0/wj8NCKmpG1foUhGiyjO3V1T2u9HwACK0cK9FNNqOY6gmH57sHTF50WZ+5aNo7g45UXgLIok9kYn6sn1v8AMij8krqeYXga4BPgFxfT0E8DrwHHtVZKmlM8G7k5Tkv8I/AfFBUaLU92/yw0qIh6iSNyPp/qyr55N54o/SnF+9AmKz/JioJKAT6C4cGcpRXJrfbHP6cClqd3DI+Jh4AyKn6dHKC70qeabFFOw96Yp4FspZiGg459NqyP5XLHZmkXSlRRXCGeNeGusOyimKB+td91mPYFHmma9nKTdVXy/ca00dXwoK4+EzSxTM++6YWbdY1OKacwhFN/lOyYi/tzckMxWT56eNTMzy+TpWTMzs0yenu1lNt5442hpaWl2GGZmq5UZM2a8EBFDq5Vz0uxlWlpamD59erPDMDNbrUh6snopT8+amZllc9I0MzPL5KRpZmaWyec0zcysV3nrrbdYsGABr7/++irb+vfvz5Zbbkm/fv06VbeTZi/z4IIXGX3iZc0Ow8ysW804751nyS9YsIBBgwbR0tLCO89ih4jgxRdfZMGCBWy99dadasfTs2Zm1qu8/vrrDBkyZKWECSCJIUOGtDkCzeWkaWZmvU7rhFltfS4nTTMzs0xOmmZmZpmcNM3MrNdp72EkXX1IiZOmmZn1Kv379+fFF19cJUFWrp7t379/p+v2V07MzKxX2XLLLVmwYAELFy5cZVvle5qd5aRpZma9Sr9+/Tr9PcxqPD1rZmaWyUnTzMwsk5OmmZlZJidNMzOzTE6aZmZmmZw0zczMMjlpmpmZZXLSNDMzy+SkaWZmlqnHJ01J4yVtnlFukqSxnaj/i5KOamN9i6S5aXmUpINL206XdEJG3ZL0B0nr1xpXG3XdKmmjrtZjZmad1+OTJjAeqJo0OysiLoqIy6oUGwUcXKVMWw4GZkXEkk7s29ovgGPrUI+ZmXVStybNNHp7SNKlkmZLukrSumnbaElTJc2QdJOkzdLIcQxwhaSZkgZIOk3SNElzJU1QB4/hlvQuSTPS8khJIWlYev+YpHXLo8YUwyxJfwS+lNatDZwBHJFiOCJVv7OkKZIel3R8OyGMA/63FM9Rqd+zJP0irZsk6UJJt6e69pd0iaQHJU0q1XUt8KkaD7mZmdVRM0aaOwITImIEsAQ4VlI/4MfA2IgYDVwCnB0RVwHTgXERMSoilgEXRMTuETEcGAB8pL2GIuJ5oH+aHt031bWvpK2A5yPitVa7/Bw4PiL2LNXxJnAacGWK4cq0aSfgg8AewHdSH1rbG6gk7V2AbwHvi4iRwFdK5TYC3gd8DbgO+CGwC7CrpFEpjpeBdSQNaa+/ZmbWWM1Imk9HxN1p+XJgH4pEOhy4RdJM4FSgvWe3HCjpT5LmUCSaXaq0dw9F8toPOCf9uy9wZ7mQpA2ADSNialr1iyr1Xh8Rb0TEC8DzwCZtlBkcEUvT8vuAq1J5IuKlUrnronjw2xzguYiYExErgAeAllK552ljqlrS0ZKmS5q+/LWlrTebmVmdNOPRYK0fmx2AgAfKI7y2SOoP/BQYExFPSzodqPY00TspkuRWFFOl30xtTm5dfRuxdeSN0vLbtH0sl0taKyXAjuqv1LWiVb0rWtXbH1jWeueImABMAFhv06279lhyMzNrVzNGmsMkVZLjp4C7gHnA0Mp6Sf3SdCbAUmBQWq4kyBckDQRyrpa9A/gX4JGUvF6iuEDn7nKhiFgELJa0T1o1rrS5HEMt5gHbpOXbgMMr06uSBtdSUTp3uykwvxNxmJlZHTQjaT4IfEbSbGAwcGE6bzgWOFfSLGAmsFcqPwm4KE3bvgFMpJjGvAaYVq2xiJifFu9I/94FLErnCFv7LPCTdCFQeUR3O8WFP+ULgXJcDxyQ4ngAOBuYmvr4gxrqARgN3BsRy2vcz8zM6kTFqbRuakxqASani3h6PUmbAZdFxAfqUNd/AtdGxG0dlVtv061jpyP/o6vNmZmtVmact8rX7WsiaUZEjKlWbnX4nuZqKyKeBSbW4+YGwNxqCdPMzBqrWy8ESlOla8QosyIifl2neibWox4zM+s8jzTNzMwyOWmamZllctI0MzPL5KRpZmaWyUnTzMwsk5OmmZlZJidNMzOzTE6aZmZmmZw0zczMMjlpmpmZZXLSNDMzy+SkaWZmlslJ08zMLJOTppmZWaZufTSYNd4/bDmE6V18GKuZmbXNI00zM7NMTppmZmaZnDTNzMwyOWmamZllctI0MzPL5KRpZmaWyUnTzMwsk5OmmZlZJidNMzOzTE6aZmZmmXwbvV7mzWcf4Kkzdm12GGZmDTHstDlNbd8jTTMzs0xOmmZmZpmcNM3MzDI5aZqZmWVy0jQzM8vkpGlmZpbJSdPMzCyTk6aZmVkmJ00zM7NMTppmZmaZnDTNzMwyOWmamZllctI0MzPL5KRpZmaWyUnTzMwsk5OmmZlZJidNMzOzTE6aZmZmmXpc0pQ0XtLmGeUmSRqbu74OcZ1SWm6RNDdzv69KOqoO7X9Z0me7Wo+ZmXVej0uawHigatJsglOqF1mZpL7A54Bf1qH9S4Dj61CPmZl1UkOTZhqRPSTpUkmzJV0lad20bbSkqZJmSLpJ0mZphDgGuELSTEkDJJ0maZqkuZImSFIN7a/SRlo/RdK5ku6T9LCkfdP6dSX9OsV6paQ/SRoj6XvAgBTTFan6PpImSnpA0s2SBrQRwvuA+yNieap/O0m3Spol6X5J20o6IMX46xTL9ySNS7HNkbQtQES8BsyXtEcnPw4zM+ui7hhp7ghMiIgRwBLgWEn9gB8DYyNiNMUo6uyIuAqYDoyLiFERsQy4ICJ2j4jhwADgIzmNttdGqUjfiNgD+CrwnbTuWODlFOuZwGiAiDgZWJZiGpfKbg/8JCJ2ARYBH28jjL2BGaX3V6R9RgJ7Ac+m9SOBrwC7AkcCO6TYLgaOK+0/Hdi3jb4eLWm6pOkvvfp2lSNjZmad1bcb2ng6Iu5Oy5dTTDHeCAwHbkkDxz68k0BaO1DSScC6wGDgAeC6jHZ3rNLG79K/M4CWtLwP8J8AETFX0uwO6n8iIma2UUfZZsCDAJIGAVtExNWp/tfTeoBpEfFsev8YcHPafw5wYKm+54GdWjcSEROACQAjthgQHcRsZmZd0B1Js/Uv8QAEPBARe3a0o6T+wE+BMRHxtKTTgf6Z7VZr443079u8cxyyp35L+1fqaGt6dhnvxNtR3eW6VpTer2Dlz6h/qtPMzJqgO6Znh0mqJK5PAXcB84ChlfWS+knaJZVZCgxKy5WE84KkgUAtV8V21EZ77gIOT+V3ppgurXgrTfnW4kFgO4CIWAIskHRYqn+dyvndGuwAZF21a2Zm9dcdSfNB4DNpqnMwcGFEvEmRAM+VNAuYSXGOD2AScJGkmRQjrokU05TXANNyG63SRnt+SpFoZwPfBGYDi9O2CcDs0oVAOW4A9iu9PxI4PtV/D7BpDXVBcY701hr3MTOzOlFE406BSWoBJqeLeHo8SX2AfhHxerpq9TaKi3Le7EKdVwMnRcQjXYxtN+DrEXFkR+VGbDEgJv/bdl1pysysxxp22pyG1CtpRkSMqVauO85prk7WBW5P07ACjulKwkxOprggqEtJE9gY+HYX6zAzsy5oaNKMiPkUV7CuFiJiKcX3ROtZ5zyK86tdreeWOoRjZmZdUPM5TUkbSRrRiGDMzMx6sqykme6gs76kwcAs4OeSftDY0MzMzHqW3JHmBukrE/8P+Hm6w85BjQvLzMys58lNmn3TfVsPByY3MB4zM7MeKzdpngHcBDwWEdMkbUPXrwY1MzNbrWRdPRsRvwF+U3r/OG3foNzMzKzXyr0QaAdJt1UevCxphKRTGxuamZlZz5I7PTsR+HfgLYCImA18slFBmZmZ9US5SXPdiLiv1brl9Q7GzMysJ8tNmi+ke7EGgKSxtP/8SzMzs14p9zZ6X6J4ysdOkv4KPAGMa1hUZmZmPVDVpClpLYqHQB8kaT1grXSPVjMzszVK1enZiFgBfDktv+qEaWZma6rcc5q3SDpB0rslDa68GhqZmZlZD5P1EGpJT7SxOiJim/qHZF0xZsyYmD59erPDMDNbrdT1IdQRsXXXQzIzM1u9ZSVNSUe1tT4iLqtvOGZmZj1X7ldOdi8t9wfeD9wPOGmamdkaI3d69rjye0kbAL9oSERmZmY9VO7Vs629Bmxfz0DMzMx6utxzmteRbqFHkWh3pvSoMDMzszVB7jnN80vLy4EnI2JBA+IxMzPrsXKnZw+OiKnpdXdELJB0bkMjMzMz62Fyk+YH2lj34XoGYmZm1tN1OD0r6RjgWGAbSbNLmwYBdzcyMDMzs56mw9vopa+WbAR8Fzi5tGlpRLzU4NisEwYOGxgjTxy50rq7j/PfN2ZmHanLbfQiYjGwGPhUqvRdFDc3GChpYEQ8VY9gzczMVgdZ5zQlfVTSIxQPn54KzAduaGBcZmZmPU7uhUBnAf8IPJxu3v5+fE7TzMzWMLlJ862IeBFYS9JaEXE7MKqBcZmZmfU4uTc3WCRpIHAncIWk5ylucmBmZrbGyB1pHkpxv9mvAjcCjwEfbVRQZmZmPVHuU05elbQVsH1EXCppXaBPY0MzMzPrWXKvnv0CcBXws7RqC+CaRgVlZmbWE+VOz34J2BtYAhARjwDvalRQZmZmPVFu0nwjIt6svJHUl3ceFWZmZrZGyE2aUyWdAgyQ9AGKZ2le17iwzMzMep7cpHkysBCYA/wb8Hvg1EYFZWZm1hNVe8rJsIh4KiJWABPTy8zMbI1UbaT59ytkJf22wbGYmZn1aNWSpkrL2zQyEDMzs56uWtKMdpbNzMzWONXuCDRS0hKKEeeAtEx6HxGxfkOjMzMz60E6HGlGRJ+IWD8iBkVE37Rced+0hCnpAEmTc9fXob3DJO1cej9FUtUnfEvarB7xSBoq6cau1mNmZl2T+5WTNd1hwM5VS63q69ThiuOIWAg8K2nvrtZlZmad15CkKWk9SddLmiVprqQj0vrRkqZKmiHpJkmbpfVTJP1I0j2p/B5p/R5p3Z/TvzvWGMMlkqal/Q9N68dL+p2kGyU9Iun7pX0+L+nhFM9ESRdI2gs4BDhP0kxJ26bin5B0Xyq/bzthfJziqTBI6iPpfElzJM2WdFxaP1/SOZL+KGm6pPekY/OYpC+W6roGGJfbfzMzq7/c52nW6kPAMxHxzwCSNpDUD/gxcGhELEyJ9Gzgc2mf9SJiL0n7AZcAw4GHgP0iYrmkg4BzKBJRjm8Bf4iIz0naELhP0q1p2yhgN+ANYJ6kHwNvA98G3gMsBf4AzIqIeyRdC0yOiKtSfwD6RsQekg4GvgMcVG5c0tbAyxHxRlp1NLA1sFvqz+BS8acjYk9JPwQmUdzntz/wAHBRKjMdOKutjko6OtXP2hutnXl4zMysVo1KmnOA8yWdS5Fs7pQ0nCIR3pKSTh/g2dI+/wMQEXdIWj8lukHApZK2p7h6t18NMfwTcIikE9L7/sCwtHxbRCwGkPQXYCtgY2BqRLyU1v8G2KGD+n+X/p0BtLSxfTOKuyhVHARcFBHLUz9fKm27Nv07BxgYEUuBpZJel7RhRCwCngc2byuQiJgATAAYOGygr3I2M2uQhiTNiHhY0mjgYOC7km4GrgYeiIg929utjfdnArdHxMcktQBTaghDwMcjYt5KK6X3UowwK96mOA7l76TmqNRR2b+1ZRSJuhxPewmtUteKVrGtKNXdP9VpZmZN0qhzmpsDr0XE5cD5FFOe84ChkvZMZfpJ2qW0W+W85z7A4jQS3AD4a9o+vsYwbgKOUxrWStqtSvn7gP0lbZSe4lKeBl5KMeqtxcOsPAK9GfhiqptW07M5dgDm1riPmZnVUaOunt2V4hziTIpzi2elR4uNBc6VNAuYCexV2udlSfdQnMP7fFr3fYqR6t0U07m1OJNiOne2pLnpfbsi4q8U50z/BNwK/AVYnDb/CjgxXVC0bTtVtK7vVeAxSdulVRcDT6V4ZgGfrrE/BwLX17iPmZnVkSKafwpM0hTghIiY3uQ4BkbEK2k0eDVwSURc3YX6PgaMjoguPxFG0h0UF1G93FG5gcMGxsgTR6607u7j7u5q82ZmvZqkGRFR9fv3/p7myk5Po+O5wBOUbljfGSnhzu9qUJKGAj+oljDNzKyxGnX1bE0i4oBmxwAQESdUL1VznRfXoY6FdDGBm5lZ13mkaWZmlslJ08zMLJOTppmZWSYnTTMzs0xOmmZmZpmcNM3MzDI5aZqZmWVy0jQzM8vkpGlmZpbJSdPMzCyTk6aZmVkmJ00zM7NMTppmZmaZnDTNzMwy9YhHg1n97PSunfzQaTOzBvFI08zMLJOTppmZWSYnTTMzs0xOmmZmZpmcNM3MzDI5aZqZmWVy0jQzM8vkpGlmZpbJSdPMzCyTk6aZmVkmJ81eZum8eUzdb/9mh2Fm1is5aZqZmWVy0jQzM8vkpGlmZpbJSdPMzCyTk6aZmVkmJ00zM7NMTppmZmaZnDTNzMwyOWmamZllctI0MzPL5KRpZmaWyUnTzMwsk5OmmZlZJidNMzOzTE6aZmZmmZw0zczMMjlpmpmZZeo1SVPSAZImd2K/zSVd1c62KZLGpOVTSutbJM3NrP+rko6qNa426vmypM92tR4zM+u8XpM0OysinomIsRlFT6leZGWS+gKfA35Zc2CrugQ4vg71mJlZJ3Vb0pS0nqTrJc2SNFfSEWn9aElTJc2QdJOkzdL6KZJ+JOmeVH6PtH6PtO7P6d8dq7T7e0kj0vKfJZ2Wls+U9K/lUaOkAZJ+JWm2pCuBAWn994ABkmZKuiJV3UfSREkPSLpZ0oA2mn8fcH9ELE/1bCfp1nQM7pe0bRohT5X0a0kPS/qepHGS7pM0R9K2ABHxGjC/chzMzKz7dedI80PAMxExMiKGAzdK6gf8GBgbEaMpRlNnl/ZZLyL2Ao5N2wAeAvaLiN2A04BzqrR7B7CvpPWB5cDeaf0+wJ2tyh4DvBYRI1IcowEi4mRgWUSMiohxqez2wE8iYhdgEfDxNtreG5hRen9F2mcksBfwbFo/EvgKsCtwJLBDROwBXAwcV9p/OrBvlf6amVmD9O3GtuYA50s6F5gcEXdKGg4MB26RBNCHdxIJwP8ARMQdktaXtCEwCLhU0vZAAP2qtHsnxbTmE8D1wAckrQu0RMQ8SS2lsvsB/5XanC1pdgf1PhERM9PyDKCljTKbAQ8CSBoEbBERV6f6X0/rAaZFxLPp/WPAzWn/OcCBpfqeB3Zq3Yiko4GjATZZZ50OQjYzs67otqQZEQ9LGg0cDHxX0s3A1cADEbFne7u18f5M4PaI+FhKeFOqND0NGAM8DtwCbAx8gZVHgB212Z43Sstvk6ZyW1kG9E/LyqxrRen9Clb+jPqnOlcSEROACQA7DhqUG7+ZmdWoO89pbk4x9Xk5cD7wHmAeMFTSnqlMP0m7lHarnPfcB1gcEYuBDYC/pu3jq7UbEW8CTwOHA/dSjDxPYNWpWSimcselNocDI0rb3krTybV4ENguxbEEWCDpsFT/OmnEW4sdgKyrds3MrP6685zmrsB9kmYC3wLOSgltLHCupFnATIpzfRUvS7oHuAj4fFr3fYqR6t0U07k57gSeSxfT3AlsSdtJ80JgYJqWPQm4r7RtAjC7dCFQjhsopnwrjgSOT/XfA2xaQ11QnCO9tcZ9zMysThTRM2fzJE0BToiI6c2OpSskXQ2cFBGPdLGe3YCvR8SRHZXbcdCgmLDbe9j/jqldac7MbI0iaUZEjKlWbo3/nmY3OJnigqCu2hj4dh3qMTOzTurOq2drEhEHNDuGeoiIeRTnbrtazy11CMfMzLrAI00zM7NMTppmZmaZnDTNzMwyOWmamZllctI0MzPL5KRpZmaWyUnTzMwsk5OmmZlZJidNMzOzTE6aZmZmmZw0zczMMjlpmpmZZXLSNDMzy+SkaWZmlslJs5cZtOOOfgC1mVmDOGmamZllctI0MzPL5KRpZmaWyUnTzMwsk5OmmZlZJkVEs2OwOpK0FJjX7DiaaGPghWYH0UTuv/vv/nfOVhExtFqhvp2s3HqueRExptlBNIuk6e6/+9/sOJrF/W98/z09a2ZmlslJ08zMLJOTZu8zodkBNJn7v2Zz/9dsDe+/LwQyMzPL5JGmmZlZJidNMzOzTE6aqylJH5I0T9Kjkk5uY/s6kq5M2/8kqaX7o2ycjP7vJ+l+ScsljW1GjI2U0f+vS/qLpNmSbpO0VTPibJSM/n9R0hxJMyXdJWnnZsTZCNX6Xio3VlJI6lVfQcn47MdLWpg++5mS/rWuAUSEX6vZC+gDPAZsA6wNzAJ2blXmWOCitPxJ4Mpmx93N/W8BRgCXAWObHXMT+n8gsG5aPmYN/PzXLy0fAtzY7Li7q++p3CDgDuBeYEyz4+7mz348cEGjYvBIc/W0B/BoRDweEW8CvwIObVXmUODStHwV8H5J6sYYG6lq/yNifkTMBlY0I8AGy+n/7RHxWnp7L7BlN8fYSDn9X1J6ux7QW654zPm/D3Am8H3g9e4Mrhvk9r9hnDRXT1sAT5feL0jr2iwTEcuBxcCQbomu8XL635vV2v/PAzc0NKLuldV/SV+S9BhF8ji+m2JrtKp9l7Qb8O6ImNydgXWT3J/9j6dTE1dJenc9A3DSXD21NWJs/Zd0TpnVVW/uW47s/kv6F2AMcF5DI+peWf2PiJ9ExLbAN4FTGx5V9+iw75LWAn4IfKPbIuoQgF9EAAAFuUlEQVReOZ/9dUBLRIwAbuWdGbe6cNJcPS0Ayn89bQk8014ZSX2BDYCXuiW6xsvpf2+W1X9JBwHfAg6JiDe6KbbuUOvn/yvgsIZG1H2q9X0QMByYImk+8I/Atb3oYqCqn31EvFj6eZ8IjK5nAE6aq6dpwPaStpa0NsWFPte2KnMt8Jm0PBb4Q6Sz5L1ATv97s6r9T1N0P6NImM83IcZGyun/9qW3/ww80o3xNVKHfY+IxRGxcUS0REQLxfnsQyJienPCrbucz36z0ttDgAfrGYCfcrIaiojlkr4M3ERxNdklEfGApDOA6RFxLfDfwC8kPUoxwvxk8yKur5z+S9oduBrYCPiopP+IiF2aGHbdZH7+5wEDgd+k67+eiohDmhZ0HWX2/8tppP0W8DLv/AG5Wsvse6+V2f/jJR0CLKf43Te+njH4NnpmZmaZPD1rZmaWyUnTzMwsk5OmmZlZJidNMzOzTE6aZmZmmZw0zbqRpLdLT1+Y2Zmnz0jaUNKx9Y/u7/WPl3RBo+pvp83DmvUkEkmbSJosaVZ6MszvmxGHrR6cNM2617KIGFV6ze9EHRtSPMWmJpL6dKKthkt3rDoMaNbju84AbomIkRGxM9Du47ZypT5ZL+SkadZkkvpIOk/StHST6X9L6wemZ2Hen54NWXmaw/eAbdNI9TxJB0iaXKrvAknj0/J8SadJugv4hKRtJd0oaYakOyXtVCW2SZIulHS7pMcl7S/pEkkPSppUKveKpP+fYr1N0tC0fpSke1O/rpa0UVo/RdI5kqZS3Bv2EOC81KdtJX0hHY9Zkn4rad1SPP8l6Z4Uz9hSDCel4zRL0vfSupz+bkZxezYA0tNxOqozp09fkTQ0xT4tvfbu6FjbaqLZz0fzy6816QW8DcxMr6vTuqOBU9PyOsB0YGuKO3atn9ZvDDxKccPqFmBuqc4DgMml9xcA49PyfOCk0rbbgO3T8nspbq/YOsbxpOcRApMo7t0qikcwLQF2pfiDewYwKpULYFxaPq20/2xg/7R8BvCjtDwF+GmpzUmUnnsKDCktnwUcVyr3m9T+zhSPiQL4MHAP7zxDdHAN/f0gsAi4neJevZtXqTO3T78E9knLw4AHm/3z51fXX55CMOteyyJiVKt1/wSMKI2aNgC2pxj9nCNpP4rngm4BbNKJNq+EYuQK7MU7t9aDIklXc11EhKQ5wHMRMSfV9wBFAp+Z4rsylb8c+J2kDYANI2JqWn8pRcJbKa52DJd0FsVU9ECK26ZVXBMRK4C/SKocj4OAn0d6hmhEvJTb34i4SdI2wIcoEuWfJQ1vp85a+nQQsHOp7fUlDYqIpR3023o4J02z5hPFSOqmlVYWU6xDgdER8ZaKp1b0b2P/5ax8qqV1mVfTv2sBi9pI2tVUnhixorRced/e75Cc+3O+2sG2ScBhETErHYcD2ogH3nlUlNpoM7u/EfESxcjwl2mqe7926qym3Ke1gD0jYlmNdVgP5nOaZs13E3CMpH4AknaQtB7FiPP5lDAPBLZK5ZdSPAKq4kmKEc06aST0/rYaiYglwBOSPpHakaSRderDWhRP0wH4NHBXRCwGXpa0b1p/JDC1rZ1ZtU+DgGfTMRmX0f7NwOdK5z4H5/ZX0vtK+w0CtgWeaqfOWvp0M/DlUju1/rFiPZBHmmbNdzHFNOf9KubyFlJcTXoFcJ2k6RRToA9B8bxASXdLmgvcEBEnSvo1xbm2R4A/d9DWOOBCSacC/SjOV86qQx9eBXaRNANYDByR1n8GuCglnseBz7az/6+AiZKOp0i+3wb+RPEHwRxWTqiriIgbU1KaLulN4PfAKeT1dzRwgaTKiP3iiJgGf090revM7dPxwE8kzab4XXsH8MWO+mE9n59yYmZdJumViBjY7DjMGs3Ts2ZmZpk80jQzM8vkkaaZmVkmJ00zM7NMTppmZmaZnDTNzMwyOWmamZll+j/whTHv0ZU11gAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "#Feature importance for Random Forest Model\n",
    "# Creating a bar plot\n",
    "sns.barplot(x=gb_feature_imp, y=gb_feature_imp.index)\n",
    "# Add labels to your graph\n",
    "plt.xlabel('Feature Importance Score')\n",
    "plt.ylabel('Features')\n",
    "plt.title(\"Visualizing Important Features\")\n",
    "plt.legend()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Feature_name\n",
       "petal width (cm)     0.409116\n",
       "petal length (cm)    0.393054\n",
       "sepal length (cm)    0.140997\n",
       "sepal width (cm)     0.056833\n",
       "dtype: float64"
      ]
     },
     "execution_count": 49,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Extra Trees\n",
    "et_feature_imp = pd.Series(et.feature_importances_,index=df_feature_names.Feature_name).sort_values(ascending=False)\n",
    "et_feature_imp"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "No handles with labels found to put in legend.\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "#Feature importance for Random Forest Model\n",
    "# Creating a bar plot\n",
    "sns.barplot(x=et_feature_imp, y=et_feature_imp.index)\n",
    "# Add labels to your graph\n",
    "plt.xlabel('Feature Importance Score')\n",
    "plt.ylabel('Features')\n",
    "plt.title(\"Visualizing Important Features\")\n",
    "plt.legend()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Question 30. Did the GB or the XTrees model perform better than the Random Forest model? Do you have any reservations about the GB and/or the XTrees model? Why? What is the main difference between the RF and the GB models?-- answered in the document"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[('Random Forest',\n",
       "  RandomForestClassifier(bootstrap=True, class_weight=None, criterion='gini',\n",
       "              max_depth=2, max_features='auto', max_leaf_nodes=None,\n",
       "              min_impurity_decrease=0.0, min_impurity_split=None,\n",
       "              min_samples_leaf=1, min_samples_split=2,\n",
       "              min_weight_fraction_leaf=0.0, n_estimators=2000, n_jobs=None,\n",
       "              oob_score=False, random_state=None, verbose=0,\n",
       "              warm_start=False)),\n",
       " ('Gradient Boosting',\n",
       "  GradientBoostingClassifier(criterion='friedman_mse', init=None,\n",
       "                learning_rate=0.1, loss='deviance', max_depth=2,\n",
       "                max_features=None, max_leaf_nodes=None,\n",
       "                min_impurity_decrease=0.0, min_impurity_split=None,\n",
       "                min_samples_leaf=1, min_samples_split=2,\n",
       "                min_weight_fraction_leaf=0.0, n_estimators=2000,\n",
       "                n_iter_no_change=None, presort='auto', random_state=None,\n",
       "                subsample=1.0, tol=0.0001, validation_fraction=0.1,\n",
       "                verbose=0, warm_start=False)),\n",
       " ('Extra Trees',\n",
       "  ExtraTreesClassifier(bootstrap=False, class_weight=None, criterion='gini',\n",
       "             max_depth=2, max_features='auto', max_leaf_nodes=None,\n",
       "             min_impurity_decrease=0.0, min_impurity_split=None,\n",
       "             min_samples_leaf=1, min_samples_split=2,\n",
       "             min_weight_fraction_leaf=0.0, n_estimators=2000, n_jobs=None,\n",
       "             oob_score=False, random_state=None, verbose=0, warm_start=False))]"
      ]
     },
     "execution_count": 51,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Accuracy comparision plot between Gradient Boosting, Xtrees and Random Forest model\n",
    "models=[]\n",
    "models.append(('Random Forest', rf))\n",
    "models.append(('Gradient Boosting', gb))\n",
    "models.append(('Extra Trees',et))\n",
    "models"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "A list of each algorithm, the mean accuracy and the standard deviation accuracy.\n",
      "Random Forest: 0.933333 (0.073030)\n",
      "Gradient Boosting: 0.933333 (0.073030)\n",
      "Extra Trees: 0.873333 (0.124544)\n"
     ]
    }
   ],
   "source": [
    "results = []\n",
    "names = []\n",
    "scoring = 'accuracy'\n",
    "print('A list of each algorithm, the mean accuracy and the standard deviation accuracy.')\n",
    "for name, model in models:\n",
    "    kfold = model_selection.KFold(n_splits=10, random_state=7)\n",
    "    cv_results = model_selection.cross_val_score(model, X, Y, cv=kfold, scoring=scoring)\n",
    "    results.append(cv_results)\n",
    "    names.append(name)\n",
    "    msg = \"%s: %f (%f)\" % (name, cv_results.mean(), cv_results.std())\n",
    "    print(msg)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# boxplot algorithm comparison\n",
    "fig = plt.figure()\n",
    "fig.suptitle('Algorithm Comparison')\n",
    "ax = fig.add_subplot(111)\n",
    "plt.boxplot(results)\n",
    "ax.set_xticklabels(names)\n",
    "#ax.set_xlabel='Algorithms'\n",
    "#ax.set_ylabel='Accuracy'\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy (train) for Gradient Boosting: 86.7% \n",
      "Accuracy (train) for Extra Trees: 83.3% \n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 7 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "X = data2.data[:, 0:2]  # we only take the first two features for visualization\n",
    "y = data2.target\n",
    "\n",
    "n_features = X.shape[1]\n",
    "\n",
    "C = 10\n",
    "kernel = 1.0 * RBF([1.0, 1.0])  # for GPC\n",
    "\n",
    "# Create different classifiers.\n",
    "classifiers = {'Gradient Boosting' : GradientBoostingClassifier(n_estimators=30, max_depth=3),\n",
    "               'Extra Trees' : ExtraTreesClassifier(n_estimators=30, max_depth=3)\n",
    "}\n",
    "\n",
    "n_classifiers = len(classifiers)\n",
    "\n",
    "plt.figure(figsize=(3 * 2, n_classifiers * 2))\n",
    "plt.subplots_adjust(bottom=.2, top=.95)\n",
    "\n",
    "xx = np.linspace(3, 9, 100)\n",
    "yy = np.linspace(1, 5, 100).T\n",
    "xx, yy = np.meshgrid(xx, yy)\n",
    "Xfull = np.c_[xx.ravel(), yy.ravel()]\n",
    "\n",
    "for index, (name, classifier) in enumerate(classifiers.items()):\n",
    "    classifier.fit(X, y)\n",
    "\n",
    "    y_pred = classifier.predict(X)\n",
    "    accuracy = accuracy_score(y, y_pred)\n",
    "    print(\"Accuracy (train) for %s: %0.1f%% \" % (name, accuracy * 100))\n",
    "\n",
    "    # View probabilities:\n",
    "    probas = classifier.predict_proba(Xfull)\n",
    "    n_classes = np.unique(y_pred).size\n",
    "    for k in range(n_classes):\n",
    "        plt.subplot(n_classifiers, n_classes, index * n_classes + k + 1)\n",
    "        plt.title(df_target_names.values[k])\n",
    "        if k == 0:\n",
    "            plt.ylabel(name)\n",
    "        imshow_handle = plt.imshow(probas[:, k].reshape((100, 100)),\n",
    "                                   extent=(3, 9, 1, 5), origin='lower')\n",
    "        plt.xticks(())\n",
    "        plt.yticks(())\n",
    "        idx = (y_pred == k)\n",
    "        if idx.any():\n",
    "            plt.scatter(X[idx, 0], X[idx, 1], marker='o', c='w', edgecolor='k')\n",
    "\n",
    "ax = plt.axes([0.15, 0.04, 0.7, 0.05])\n",
    "plt.title(\"Probability\")\n",
    "plt.colorbar(imshow_handle, cax=ax, orientation='horizontal')\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.6.9"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}